Meng-Fu Bryan Tsou, PhD

The Centrosome and Human Disease — Why We Care

Centrosome biogenesis is of great interest because centrosome amplification occurs frequently in almost all types of cancer, and is considered to be a major contributing factor for genomic instability in cancer cells. A vast amount of published work has revealed that centrosome amplification strongly correlates with loss of tumor suppressor proteins such as p53 and BRCA1, and with infections of human papillomaviruses (HPV). The mechanism behind these correlations remains poorly understood because little is known about how centrosome biogenesis is controlled in normal cells. In addition to cancer, centrosome dysfunction has also been implicated in other human diseases, many related to the roles of centrioles, which make up the core of the centrosome, in the formation of cilia and flagella.

The Centrosome Duplication Cycle - The “Once-Only” Control

Somatically dividing animal cells have a single centrosome, which is usually defined as a pair of centrioles surrounded by pericentriolar material (PCM) that nucleates and organizes microtubule arrays. Each centriole is a barrel-shaped structure made of nine microtubule triplets and associated proteins. In most animal cells, by unknown mechanism, centrioles strictly determine the localization of the PCM, and therefore determine the number of centrosomes. Cells begin the cell cycle with a single centrosome containing a pair of centrioles; this centrosome duplicates once and only once so that at mitosis, there are two centrosomes (or two pairs of centrioles). The centrosomes segregate with the spindle poles during mitosis so that after cell division each daughter cell receives one centrosome with a pair of centrioles. From a mechanistic perspective, the problem of centrosome duplication comes down to the question of how centrioles are duplicated. Recently, we have developed an in vitro system using Xenopus egg extract and purified centrioles to characterize the regulation of centrosome duplication. In this system, all major cell cycle events of centrosome duplication can be recapitulated, including centriole disengagement and centriole growth. This in vitro system provides us with a powerful tool to dissect the mechanism by which centrosome duplication occurs once and only once per cell cycle.

De Novo Centriole Formation and Number Control

Although in most dividing cells centriole duplication requires a preexisting centriole (canonical centriole duplication cycle), de novo centriole formation does occur naturally in some cases, such as multi-ciliated epithelial cells and early mouse embryos. Strikingly, de novo assembly can also occur in vertebrate somatic cells when the endogenous centrioles are destroyed or removed. The number of centrioles formed through the de novo pathway is highly variable, posing a grave risk for dividing cells, which require strict control over centrosome number to maintain genomic stability. A critical question then is how the de novo assembly pathway is normally inhibited so that only the tightly-controlled canonical pathway is used.

Sex and the Single Centrosome

The canonical centriole duplication cycle described above is interrupted during sexual reproduction. Sexual reproduction requires reduction in the gametes of both chromosome number and centrosome number so that upon fertilization, the correct number of chromosomes and centrosomes is restored. For chromosomes, the problem is solved during meiosis in which the DNA content in gametes is reduced by half in both sexes. For centrosomes, however, in most animals mature sperm cells retain all or part of the centriole pair, whereas oocytes lose them entirely. The union of sperm and egg during fertilization restores the proper number of centrioles in zygotes. Despite its pivotal role in sexual reproduction, very little is known about how centrioles, which are normally very stable, are selectively degenerated during oogenesis.

Ongoing and Future Projects

Our lab focuses on the processes described above using several experimental systems, including the model organism C. elegans, which has strong genetics, and human cells and Xenopus egg extracts, which are the systems most relevant to our desire to understand the human centrosome in normal cell division and in disease. Some of the projects we are interested in include:

1. The role of separase in centrosome duplication.
2. The recruitment of pericentriolar material after centriole disengagement.
3. The control of de novo centrosome formation in somatic cycling cells.
4. The mechanism of centrosome degeneration in gametogenesis, and the role of this mechanism in somatic cells.
5. The causes of centrosome amplification in cancer cells associated with loss of p53, BRCA1, or with infections of human papillomaviruses.